Add files using upload-large-folder tool
Browse filesThis view is limited to 50 files because it contains too many changes. See raw diff
- 2b_l10/outputs_add_free/all/lang/103/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/104/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/107/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/108/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/112/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/114/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/116/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/128/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/13/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/131/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/137/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/144/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/147/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/158/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/159/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/160/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/167/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/174/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/185/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/188/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/190/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/195/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/196/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/197/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/199/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/204/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/207/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/208/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/215/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/22/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/231/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/232/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/240/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/241/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/242/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/245/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/246/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/247/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/255/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/260/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/265/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/271/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/28/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/288/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/289/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/29/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/292/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/293/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/294/config.json +12 -0
- 2b_l10/outputs_add_free/all/lang/298/config.json +12 -0
2b_l10/outputs_add_free/all/lang/103/config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[
|
| 2 |
+
{
|
| 3 |
+
"layer": 10,
|
| 4 |
+
"embed_dim": 2304,
|
| 5 |
+
"low_rank_dim": 1,
|
| 6 |
+
"target_module": "model.layers.10",
|
| 7 |
+
"intervention_type": "AdditionFreeIntervention",
|
| 8 |
+
"factor_init_scale": 2.0,
|
| 9 |
+
"vector_init_scale": 1.0,
|
| 10 |
+
"alpha": 0.0
|
| 11 |
+
}
|
| 12 |
+
]
|
2b_l10/outputs_add_free/all/lang/104/config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[
|
| 2 |
+
{
|
| 3 |
+
"layer": 10,
|
| 4 |
+
"embed_dim": 2304,
|
| 5 |
+
"low_rank_dim": 1,
|
| 6 |
+
"target_module": "model.layers.10",
|
| 7 |
+
"intervention_type": "AdditionFreeIntervention",
|
| 8 |
+
"factor_init_scale": 2.0,
|
| 9 |
+
"vector_init_scale": 1.0,
|
| 10 |
+
"alpha": 0.0
|
| 11 |
+
}
|
| 12 |
+
]
|
2b_l10/outputs_add_free/all/lang/107/config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[
|
| 2 |
+
{
|
| 3 |
+
"layer": 10,
|
| 4 |
+
"embed_dim": 2304,
|
| 5 |
+
"low_rank_dim": 1,
|
| 6 |
+
"target_module": "model.layers.10",
|
| 7 |
+
"intervention_type": "AdditionFreeIntervention",
|
| 8 |
+
"factor_init_scale": 2.0,
|
| 9 |
+
"vector_init_scale": 1.0,
|
| 10 |
+
"alpha": 0.0
|
| 11 |
+
}
|
| 12 |
+
]
|
2b_l10/outputs_add_free/all/lang/108/config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[
|
| 2 |
+
{
|
| 3 |
+
"layer": 10,
|
| 4 |
+
"embed_dim": 2304,
|
| 5 |
+
"low_rank_dim": 1,
|
| 6 |
+
"target_module": "model.layers.10",
|
| 7 |
+
"intervention_type": "AdditionFreeIntervention",
|
| 8 |
+
"factor_init_scale": 2.0,
|
| 9 |
+
"vector_init_scale": 1.0,
|
| 10 |
+
"alpha": 0.0
|
| 11 |
+
}
|
| 12 |
+
]
|
2b_l10/outputs_add_free/all/lang/112/config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[
|
| 2 |
+
{
|
| 3 |
+
"layer": 10,
|
| 4 |
+
"embed_dim": 2304,
|
| 5 |
+
"low_rank_dim": 1,
|
| 6 |
+
"target_module": "model.layers.10",
|
| 7 |
+
"intervention_type": "AdditionFreeIntervention",
|
| 8 |
+
"factor_init_scale": 2.0,
|
| 9 |
+
"vector_init_scale": 1.0,
|
| 10 |
+
"alpha": 0.0
|
| 11 |
+
}
|
| 12 |
+
]
|
2b_l10/outputs_add_free/all/lang/114/config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[
|
| 2 |
+
{
|
| 3 |
+
"layer": 10,
|
| 4 |
+
"embed_dim": 2304,
|
| 5 |
+
"low_rank_dim": 1,
|
| 6 |
+
"target_module": "model.layers.10",
|
| 7 |
+
"intervention_type": "AdditionFreeIntervention",
|
| 8 |
+
"factor_init_scale": 2.0,
|
| 9 |
+
"vector_init_scale": 1.0,
|
| 10 |
+
"alpha": 0.0
|
| 11 |
+
}
|
| 12 |
+
]
|
2b_l10/outputs_add_free/all/lang/116/config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[
|
| 2 |
+
{
|
| 3 |
+
"layer": 10,
|
| 4 |
+
"embed_dim": 2304,
|
| 5 |
+
"low_rank_dim": 1,
|
| 6 |
+
"target_module": "model.layers.10",
|
| 7 |
+
"intervention_type": "AdditionFreeIntervention",
|
| 8 |
+
"factor_init_scale": 2.0,
|
| 9 |
+
"vector_init_scale": 1.0,
|
| 10 |
+
"alpha": 0.0
|
| 11 |
+
}
|
| 12 |
+
]
|
2b_l10/outputs_add_free/all/lang/128/config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[
|
| 2 |
+
{
|
| 3 |
+
"layer": 10,
|
| 4 |
+
"embed_dim": 2304,
|
| 5 |
+
"low_rank_dim": 1,
|
| 6 |
+
"target_module": "model.layers.10",
|
| 7 |
+
"intervention_type": "AdditionFreeIntervention",
|
| 8 |
+
"factor_init_scale": 2.0,
|
| 9 |
+
"vector_init_scale": 1.0,
|
| 10 |
+
"alpha": 0.0
|
| 11 |
+
}
|
| 12 |
+
]
|
2b_l10/outputs_add_free/all/lang/13/config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[
|
| 2 |
+
{
|
| 3 |
+
"layer": 10,
|
| 4 |
+
"embed_dim": 2304,
|
| 5 |
+
"low_rank_dim": 1,
|
| 6 |
+
"target_module": "model.layers.10",
|
| 7 |
+
"intervention_type": "AdditionFreeIntervention",
|
| 8 |
+
"factor_init_scale": 2.0,
|
| 9 |
+
"vector_init_scale": 1.0,
|
| 10 |
+
"alpha": 0.0
|
| 11 |
+
}
|
| 12 |
+
]
|
2b_l10/outputs_add_free/all/lang/131/config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[
|
| 2 |
+
{
|
| 3 |
+
"layer": 10,
|
| 4 |
+
"embed_dim": 2304,
|
| 5 |
+
"low_rank_dim": 1,
|
| 6 |
+
"target_module": "model.layers.10",
|
| 7 |
+
"intervention_type": "AdditionFreeIntervention",
|
| 8 |
+
"factor_init_scale": 2.0,
|
| 9 |
+
"vector_init_scale": 1.0,
|
| 10 |
+
"alpha": 0.0
|
| 11 |
+
}
|
| 12 |
+
]
|
2b_l10/outputs_add_free/all/lang/137/config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[
|
| 2 |
+
{
|
| 3 |
+
"layer": 10,
|
| 4 |
+
"embed_dim": 2304,
|
| 5 |
+
"low_rank_dim": 1,
|
| 6 |
+
"target_module": "model.layers.10",
|
| 7 |
+
"intervention_type": "AdditionFreeIntervention",
|
| 8 |
+
"factor_init_scale": 2.0,
|
| 9 |
+
"vector_init_scale": 1.0,
|
| 10 |
+
"alpha": 0.0
|
| 11 |
+
}
|
| 12 |
+
]
|
2b_l10/outputs_add_free/all/lang/144/config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[
|
| 2 |
+
{
|
| 3 |
+
"layer": 10,
|
| 4 |
+
"embed_dim": 2304,
|
| 5 |
+
"low_rank_dim": 1,
|
| 6 |
+
"target_module": "model.layers.10",
|
| 7 |
+
"intervention_type": "AdditionFreeIntervention",
|
| 8 |
+
"factor_init_scale": 2.0,
|
| 9 |
+
"vector_init_scale": 1.0,
|
| 10 |
+
"alpha": 0.0
|
| 11 |
+
}
|
| 12 |
+
]
|
2b_l10/outputs_add_free/all/lang/147/config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[
|
| 2 |
+
{
|
| 3 |
+
"layer": 10,
|
| 4 |
+
"embed_dim": 2304,
|
| 5 |
+
"low_rank_dim": 1,
|
| 6 |
+
"target_module": "model.layers.10",
|
| 7 |
+
"intervention_type": "AdditionFreeIntervention",
|
| 8 |
+
"factor_init_scale": 2.0,
|
| 9 |
+
"vector_init_scale": 1.0,
|
| 10 |
+
"alpha": 0.0
|
| 11 |
+
}
|
| 12 |
+
]
|
2b_l10/outputs_add_free/all/lang/158/config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[
|
| 2 |
+
{
|
| 3 |
+
"layer": 10,
|
| 4 |
+
"embed_dim": 2304,
|
| 5 |
+
"low_rank_dim": 1,
|
| 6 |
+
"target_module": "model.layers.10",
|
| 7 |
+
"intervention_type": "AdditionFreeIntervention",
|
| 8 |
+
"factor_init_scale": 2.0,
|
| 9 |
+
"vector_init_scale": 1.0,
|
| 10 |
+
"alpha": 0.0
|
| 11 |
+
}
|
| 12 |
+
]
|
2b_l10/outputs_add_free/all/lang/159/config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[
|
| 2 |
+
{
|
| 3 |
+
"layer": 10,
|
| 4 |
+
"embed_dim": 2304,
|
| 5 |
+
"low_rank_dim": 1,
|
| 6 |
+
"target_module": "model.layers.10",
|
| 7 |
+
"intervention_type": "AdditionFreeIntervention",
|
| 8 |
+
"factor_init_scale": 2.0,
|
| 9 |
+
"vector_init_scale": 1.0,
|
| 10 |
+
"alpha": 0.0
|
| 11 |
+
}
|
| 12 |
+
]
|
2b_l10/outputs_add_free/all/lang/160/config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[
|
| 2 |
+
{
|
| 3 |
+
"layer": 10,
|
| 4 |
+
"embed_dim": 2304,
|
| 5 |
+
"low_rank_dim": 1,
|
| 6 |
+
"target_module": "model.layers.10",
|
| 7 |
+
"intervention_type": "AdditionFreeIntervention",
|
| 8 |
+
"factor_init_scale": 2.0,
|
| 9 |
+
"vector_init_scale": 1.0,
|
| 10 |
+
"alpha": 0.0
|
| 11 |
+
}
|
| 12 |
+
]
|
2b_l10/outputs_add_free/all/lang/167/config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[
|
| 2 |
+
{
|
| 3 |
+
"layer": 10,
|
| 4 |
+
"embed_dim": 2304,
|
| 5 |
+
"low_rank_dim": 1,
|
| 6 |
+
"target_module": "model.layers.10",
|
| 7 |
+
"intervention_type": "AdditionFreeIntervention",
|
| 8 |
+
"factor_init_scale": 2.0,
|
| 9 |
+
"vector_init_scale": 1.0,
|
| 10 |
+
"alpha": 0.0
|
| 11 |
+
}
|
| 12 |
+
]
|
2b_l10/outputs_add_free/all/lang/174/config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[
|
| 2 |
+
{
|
| 3 |
+
"layer": 10,
|
| 4 |
+
"embed_dim": 2304,
|
| 5 |
+
"low_rank_dim": 1,
|
| 6 |
+
"target_module": "model.layers.10",
|
| 7 |
+
"intervention_type": "AdditionFreeIntervention",
|
| 8 |
+
"factor_init_scale": 2.0,
|
| 9 |
+
"vector_init_scale": 1.0,
|
| 10 |
+
"alpha": 0.0
|
| 11 |
+
}
|
| 12 |
+
]
|
2b_l10/outputs_add_free/all/lang/185/config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[
|
| 2 |
+
{
|
| 3 |
+
"layer": 10,
|
| 4 |
+
"embed_dim": 2304,
|
| 5 |
+
"low_rank_dim": 1,
|
| 6 |
+
"target_module": "model.layers.10",
|
| 7 |
+
"intervention_type": "AdditionFreeIntervention",
|
| 8 |
+
"factor_init_scale": 2.0,
|
| 9 |
+
"vector_init_scale": 1.0,
|
| 10 |
+
"alpha": 0.0
|
| 11 |
+
}
|
| 12 |
+
]
|
2b_l10/outputs_add_free/all/lang/188/config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[
|
| 2 |
+
{
|
| 3 |
+
"layer": 10,
|
| 4 |
+
"embed_dim": 2304,
|
| 5 |
+
"low_rank_dim": 1,
|
| 6 |
+
"target_module": "model.layers.10",
|
| 7 |
+
"intervention_type": "AdditionFreeIntervention",
|
| 8 |
+
"factor_init_scale": 2.0,
|
| 9 |
+
"vector_init_scale": 1.0,
|
| 10 |
+
"alpha": 0.0
|
| 11 |
+
}
|
| 12 |
+
]
|
2b_l10/outputs_add_free/all/lang/190/config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[
|
| 2 |
+
{
|
| 3 |
+
"layer": 10,
|
| 4 |
+
"embed_dim": 2304,
|
| 5 |
+
"low_rank_dim": 1,
|
| 6 |
+
"target_module": "model.layers.10",
|
| 7 |
+
"intervention_type": "AdditionFreeIntervention",
|
| 8 |
+
"factor_init_scale": 2.0,
|
| 9 |
+
"vector_init_scale": 1.0,
|
| 10 |
+
"alpha": 0.0
|
| 11 |
+
}
|
| 12 |
+
]
|
2b_l10/outputs_add_free/all/lang/195/config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[
|
| 2 |
+
{
|
| 3 |
+
"layer": 10,
|
| 4 |
+
"embed_dim": 2304,
|
| 5 |
+
"low_rank_dim": 1,
|
| 6 |
+
"target_module": "model.layers.10",
|
| 7 |
+
"intervention_type": "AdditionFreeIntervention",
|
| 8 |
+
"factor_init_scale": 2.0,
|
| 9 |
+
"vector_init_scale": 1.0,
|
| 10 |
+
"alpha": 0.0
|
| 11 |
+
}
|
| 12 |
+
]
|
2b_l10/outputs_add_free/all/lang/196/config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[
|
| 2 |
+
{
|
| 3 |
+
"layer": 10,
|
| 4 |
+
"embed_dim": 2304,
|
| 5 |
+
"low_rank_dim": 1,
|
| 6 |
+
"target_module": "model.layers.10",
|
| 7 |
+
"intervention_type": "AdditionFreeIntervention",
|
| 8 |
+
"factor_init_scale": 2.0,
|
| 9 |
+
"vector_init_scale": 1.0,
|
| 10 |
+
"alpha": 0.0
|
| 11 |
+
}
|
| 12 |
+
]
|
2b_l10/outputs_add_free/all/lang/197/config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[
|
| 2 |
+
{
|
| 3 |
+
"layer": 10,
|
| 4 |
+
"embed_dim": 2304,
|
| 5 |
+
"low_rank_dim": 1,
|
| 6 |
+
"target_module": "model.layers.10",
|
| 7 |
+
"intervention_type": "AdditionFreeIntervention",
|
| 8 |
+
"factor_init_scale": 2.0,
|
| 9 |
+
"vector_init_scale": 1.0,
|
| 10 |
+
"alpha": 0.0
|
| 11 |
+
}
|
| 12 |
+
]
|
2b_l10/outputs_add_free/all/lang/199/config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[
|
| 2 |
+
{
|
| 3 |
+
"layer": 10,
|
| 4 |
+
"embed_dim": 2304,
|
| 5 |
+
"low_rank_dim": 1,
|
| 6 |
+
"target_module": "model.layers.10",
|
| 7 |
+
"intervention_type": "AdditionFreeIntervention",
|
| 8 |
+
"factor_init_scale": 2.0,
|
| 9 |
+
"vector_init_scale": 1.0,
|
| 10 |
+
"alpha": 0.0
|
| 11 |
+
}
|
| 12 |
+
]
|
2b_l10/outputs_add_free/all/lang/204/config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[
|
| 2 |
+
{
|
| 3 |
+
"layer": 10,
|
| 4 |
+
"embed_dim": 2304,
|
| 5 |
+
"low_rank_dim": 1,
|
| 6 |
+
"target_module": "model.layers.10",
|
| 7 |
+
"intervention_type": "AdditionFreeIntervention",
|
| 8 |
+
"factor_init_scale": 2.0,
|
| 9 |
+
"vector_init_scale": 1.0,
|
| 10 |
+
"alpha": 0.0
|
| 11 |
+
}
|
| 12 |
+
]
|
2b_l10/outputs_add_free/all/lang/207/config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[
|
| 2 |
+
{
|
| 3 |
+
"layer": 10,
|
| 4 |
+
"embed_dim": 2304,
|
| 5 |
+
"low_rank_dim": 1,
|
| 6 |
+
"target_module": "model.layers.10",
|
| 7 |
+
"intervention_type": "AdditionFreeIntervention",
|
| 8 |
+
"factor_init_scale": 2.0,
|
| 9 |
+
"vector_init_scale": 1.0,
|
| 10 |
+
"alpha": 0.0
|
| 11 |
+
}
|
| 12 |
+
]
|
2b_l10/outputs_add_free/all/lang/208/config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[
|
| 2 |
+
{
|
| 3 |
+
"layer": 10,
|
| 4 |
+
"embed_dim": 2304,
|
| 5 |
+
"low_rank_dim": 1,
|
| 6 |
+
"target_module": "model.layers.10",
|
| 7 |
+
"intervention_type": "AdditionFreeIntervention",
|
| 8 |
+
"factor_init_scale": 2.0,
|
| 9 |
+
"vector_init_scale": 1.0,
|
| 10 |
+
"alpha": 0.0
|
| 11 |
+
}
|
| 12 |
+
]
|
2b_l10/outputs_add_free/all/lang/215/config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[
|
| 2 |
+
{
|
| 3 |
+
"layer": 10,
|
| 4 |
+
"embed_dim": 2304,
|
| 5 |
+
"low_rank_dim": 1,
|
| 6 |
+
"target_module": "model.layers.10",
|
| 7 |
+
"intervention_type": "AdditionFreeIntervention",
|
| 8 |
+
"factor_init_scale": 2.0,
|
| 9 |
+
"vector_init_scale": 1.0,
|
| 10 |
+
"alpha": 0.0
|
| 11 |
+
}
|
| 12 |
+
]
|
2b_l10/outputs_add_free/all/lang/22/config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[
|
| 2 |
+
{
|
| 3 |
+
"layer": 10,
|
| 4 |
+
"embed_dim": 2304,
|
| 5 |
+
"low_rank_dim": 1,
|
| 6 |
+
"target_module": "model.layers.10",
|
| 7 |
+
"intervention_type": "AdditionFreeIntervention",
|
| 8 |
+
"factor_init_scale": 2.0,
|
| 9 |
+
"vector_init_scale": 1.0,
|
| 10 |
+
"alpha": 0.0
|
| 11 |
+
}
|
| 12 |
+
]
|
2b_l10/outputs_add_free/all/lang/231/config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[
|
| 2 |
+
{
|
| 3 |
+
"layer": 10,
|
| 4 |
+
"embed_dim": 2304,
|
| 5 |
+
"low_rank_dim": 1,
|
| 6 |
+
"target_module": "model.layers.10",
|
| 7 |
+
"intervention_type": "AdditionFreeIntervention",
|
| 8 |
+
"factor_init_scale": 2.0,
|
| 9 |
+
"vector_init_scale": 1.0,
|
| 10 |
+
"alpha": 0.0
|
| 11 |
+
}
|
| 12 |
+
]
|
2b_l10/outputs_add_free/all/lang/232/config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[
|
| 2 |
+
{
|
| 3 |
+
"layer": 10,
|
| 4 |
+
"embed_dim": 2304,
|
| 5 |
+
"low_rank_dim": 1,
|
| 6 |
+
"target_module": "model.layers.10",
|
| 7 |
+
"intervention_type": "AdditionFreeIntervention",
|
| 8 |
+
"factor_init_scale": 2.0,
|
| 9 |
+
"vector_init_scale": 1.0,
|
| 10 |
+
"alpha": 0.0
|
| 11 |
+
}
|
| 12 |
+
]
|
2b_l10/outputs_add_free/all/lang/240/config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[
|
| 2 |
+
{
|
| 3 |
+
"layer": 10,
|
| 4 |
+
"embed_dim": 2304,
|
| 5 |
+
"low_rank_dim": 1,
|
| 6 |
+
"target_module": "model.layers.10",
|
| 7 |
+
"intervention_type": "AdditionFreeIntervention",
|
| 8 |
+
"factor_init_scale": 2.0,
|
| 9 |
+
"vector_init_scale": 1.0,
|
| 10 |
+
"alpha": 0.0
|
| 11 |
+
}
|
| 12 |
+
]
|
2b_l10/outputs_add_free/all/lang/241/config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[
|
| 2 |
+
{
|
| 3 |
+
"layer": 10,
|
| 4 |
+
"embed_dim": 2304,
|
| 5 |
+
"low_rank_dim": 1,
|
| 6 |
+
"target_module": "model.layers.10",
|
| 7 |
+
"intervention_type": "AdditionFreeIntervention",
|
| 8 |
+
"factor_init_scale": 2.0,
|
| 9 |
+
"vector_init_scale": 1.0,
|
| 10 |
+
"alpha": 0.0
|
| 11 |
+
}
|
| 12 |
+
]
|
2b_l10/outputs_add_free/all/lang/242/config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[
|
| 2 |
+
{
|
| 3 |
+
"layer": 10,
|
| 4 |
+
"embed_dim": 2304,
|
| 5 |
+
"low_rank_dim": 1,
|
| 6 |
+
"target_module": "model.layers.10",
|
| 7 |
+
"intervention_type": "AdditionFreeIntervention",
|
| 8 |
+
"factor_init_scale": 2.0,
|
| 9 |
+
"vector_init_scale": 1.0,
|
| 10 |
+
"alpha": 0.0
|
| 11 |
+
}
|
| 12 |
+
]
|
2b_l10/outputs_add_free/all/lang/245/config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[
|
| 2 |
+
{
|
| 3 |
+
"layer": 10,
|
| 4 |
+
"embed_dim": 2304,
|
| 5 |
+
"low_rank_dim": 1,
|
| 6 |
+
"target_module": "model.layers.10",
|
| 7 |
+
"intervention_type": "AdditionFreeIntervention",
|
| 8 |
+
"factor_init_scale": 2.0,
|
| 9 |
+
"vector_init_scale": 1.0,
|
| 10 |
+
"alpha": 0.0
|
| 11 |
+
}
|
| 12 |
+
]
|
2b_l10/outputs_add_free/all/lang/246/config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[
|
| 2 |
+
{
|
| 3 |
+
"layer": 10,
|
| 4 |
+
"embed_dim": 2304,
|
| 5 |
+
"low_rank_dim": 1,
|
| 6 |
+
"target_module": "model.layers.10",
|
| 7 |
+
"intervention_type": "AdditionFreeIntervention",
|
| 8 |
+
"factor_init_scale": 2.0,
|
| 9 |
+
"vector_init_scale": 1.0,
|
| 10 |
+
"alpha": 0.0
|
| 11 |
+
}
|
| 12 |
+
]
|
2b_l10/outputs_add_free/all/lang/247/config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[
|
| 2 |
+
{
|
| 3 |
+
"layer": 10,
|
| 4 |
+
"embed_dim": 2304,
|
| 5 |
+
"low_rank_dim": 1,
|
| 6 |
+
"target_module": "model.layers.10",
|
| 7 |
+
"intervention_type": "AdditionFreeIntervention",
|
| 8 |
+
"factor_init_scale": 2.0,
|
| 9 |
+
"vector_init_scale": 1.0,
|
| 10 |
+
"alpha": 0.0
|
| 11 |
+
}
|
| 12 |
+
]
|
2b_l10/outputs_add_free/all/lang/255/config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[
|
| 2 |
+
{
|
| 3 |
+
"layer": 10,
|
| 4 |
+
"embed_dim": 2304,
|
| 5 |
+
"low_rank_dim": 1,
|
| 6 |
+
"target_module": "model.layers.10",
|
| 7 |
+
"intervention_type": "AdditionFreeIntervention",
|
| 8 |
+
"factor_init_scale": 2.0,
|
| 9 |
+
"vector_init_scale": 1.0,
|
| 10 |
+
"alpha": 0.0
|
| 11 |
+
}
|
| 12 |
+
]
|
2b_l10/outputs_add_free/all/lang/260/config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[
|
| 2 |
+
{
|
| 3 |
+
"layer": 10,
|
| 4 |
+
"embed_dim": 2304,
|
| 5 |
+
"low_rank_dim": 1,
|
| 6 |
+
"target_module": "model.layers.10",
|
| 7 |
+
"intervention_type": "AdditionFreeIntervention",
|
| 8 |
+
"factor_init_scale": 2.0,
|
| 9 |
+
"vector_init_scale": 1.0,
|
| 10 |
+
"alpha": 0.0
|
| 11 |
+
}
|
| 12 |
+
]
|
2b_l10/outputs_add_free/all/lang/265/config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[
|
| 2 |
+
{
|
| 3 |
+
"layer": 10,
|
| 4 |
+
"embed_dim": 2304,
|
| 5 |
+
"low_rank_dim": 1,
|
| 6 |
+
"target_module": "model.layers.10",
|
| 7 |
+
"intervention_type": "AdditionFreeIntervention",
|
| 8 |
+
"factor_init_scale": 2.0,
|
| 9 |
+
"vector_init_scale": 1.0,
|
| 10 |
+
"alpha": 0.0
|
| 11 |
+
}
|
| 12 |
+
]
|
2b_l10/outputs_add_free/all/lang/271/config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[
|
| 2 |
+
{
|
| 3 |
+
"layer": 10,
|
| 4 |
+
"embed_dim": 2304,
|
| 5 |
+
"low_rank_dim": 1,
|
| 6 |
+
"target_module": "model.layers.10",
|
| 7 |
+
"intervention_type": "AdditionFreeIntervention",
|
| 8 |
+
"factor_init_scale": 2.0,
|
| 9 |
+
"vector_init_scale": 1.0,
|
| 10 |
+
"alpha": 0.0
|
| 11 |
+
}
|
| 12 |
+
]
|
2b_l10/outputs_add_free/all/lang/28/config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[
|
| 2 |
+
{
|
| 3 |
+
"layer": 10,
|
| 4 |
+
"embed_dim": 2304,
|
| 5 |
+
"low_rank_dim": 1,
|
| 6 |
+
"target_module": "model.layers.10",
|
| 7 |
+
"intervention_type": "AdditionFreeIntervention",
|
| 8 |
+
"factor_init_scale": 2.0,
|
| 9 |
+
"vector_init_scale": 1.0,
|
| 10 |
+
"alpha": 0.0
|
| 11 |
+
}
|
| 12 |
+
]
|
2b_l10/outputs_add_free/all/lang/288/config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[
|
| 2 |
+
{
|
| 3 |
+
"layer": 10,
|
| 4 |
+
"embed_dim": 2304,
|
| 5 |
+
"low_rank_dim": 1,
|
| 6 |
+
"target_module": "model.layers.10",
|
| 7 |
+
"intervention_type": "AdditionFreeIntervention",
|
| 8 |
+
"factor_init_scale": 2.0,
|
| 9 |
+
"vector_init_scale": 1.0,
|
| 10 |
+
"alpha": 0.0
|
| 11 |
+
}
|
| 12 |
+
]
|
2b_l10/outputs_add_free/all/lang/289/config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[
|
| 2 |
+
{
|
| 3 |
+
"layer": 10,
|
| 4 |
+
"embed_dim": 2304,
|
| 5 |
+
"low_rank_dim": 1,
|
| 6 |
+
"target_module": "model.layers.10",
|
| 7 |
+
"intervention_type": "AdditionFreeIntervention",
|
| 8 |
+
"factor_init_scale": 2.0,
|
| 9 |
+
"vector_init_scale": 1.0,
|
| 10 |
+
"alpha": 0.0
|
| 11 |
+
}
|
| 12 |
+
]
|
2b_l10/outputs_add_free/all/lang/29/config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[
|
| 2 |
+
{
|
| 3 |
+
"layer": 10,
|
| 4 |
+
"embed_dim": 2304,
|
| 5 |
+
"low_rank_dim": 1,
|
| 6 |
+
"target_module": "model.layers.10",
|
| 7 |
+
"intervention_type": "AdditionFreeIntervention",
|
| 8 |
+
"factor_init_scale": 2.0,
|
| 9 |
+
"vector_init_scale": 1.0,
|
| 10 |
+
"alpha": 0.0
|
| 11 |
+
}
|
| 12 |
+
]
|
2b_l10/outputs_add_free/all/lang/292/config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[
|
| 2 |
+
{
|
| 3 |
+
"layer": 10,
|
| 4 |
+
"embed_dim": 2304,
|
| 5 |
+
"low_rank_dim": 1,
|
| 6 |
+
"target_module": "model.layers.10",
|
| 7 |
+
"intervention_type": "AdditionFreeIntervention",
|
| 8 |
+
"factor_init_scale": 2.0,
|
| 9 |
+
"vector_init_scale": 1.0,
|
| 10 |
+
"alpha": 0.0
|
| 11 |
+
}
|
| 12 |
+
]
|
2b_l10/outputs_add_free/all/lang/293/config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[
|
| 2 |
+
{
|
| 3 |
+
"layer": 10,
|
| 4 |
+
"embed_dim": 2304,
|
| 5 |
+
"low_rank_dim": 1,
|
| 6 |
+
"target_module": "model.layers.10",
|
| 7 |
+
"intervention_type": "AdditionFreeIntervention",
|
| 8 |
+
"factor_init_scale": 2.0,
|
| 9 |
+
"vector_init_scale": 1.0,
|
| 10 |
+
"alpha": 0.0
|
| 11 |
+
}
|
| 12 |
+
]
|
2b_l10/outputs_add_free/all/lang/294/config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[
|
| 2 |
+
{
|
| 3 |
+
"layer": 10,
|
| 4 |
+
"embed_dim": 2304,
|
| 5 |
+
"low_rank_dim": 1,
|
| 6 |
+
"target_module": "model.layers.10",
|
| 7 |
+
"intervention_type": "AdditionFreeIntervention",
|
| 8 |
+
"factor_init_scale": 2.0,
|
| 9 |
+
"vector_init_scale": 1.0,
|
| 10 |
+
"alpha": 0.0
|
| 11 |
+
}
|
| 12 |
+
]
|
2b_l10/outputs_add_free/all/lang/298/config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[
|
| 2 |
+
{
|
| 3 |
+
"layer": 10,
|
| 4 |
+
"embed_dim": 2304,
|
| 5 |
+
"low_rank_dim": 1,
|
| 6 |
+
"target_module": "model.layers.10",
|
| 7 |
+
"intervention_type": "AdditionFreeIntervention",
|
| 8 |
+
"factor_init_scale": 2.0,
|
| 9 |
+
"vector_init_scale": 1.0,
|
| 10 |
+
"alpha": 0.0
|
| 11 |
+
}
|
| 12 |
+
]
|